Reliable scheduling for long-running agent work
Durable queues, leases, checkpoints, and halt states turn fragile agent loops into trusted operations.
Long-running agent work fails differently than web requests. The scheduler must expect timeouts, rate limits, approvals, deploys, and resumption.
AstraRunner treats this as an execution-platform concern: the run needs durable state, traceable decisions, cost visibility, and enough structure for teams to review what happened after the agent finishes.
Patterns that work
- Persist every step before execution.
- Use leases so one worker owns a step.
- Checkpoint before side effects.
- Expose waiting, blocked, failed, and complete states.
Implementation checklist
- Define retry policy per failure type.
- Store external operation IDs.
- Make resume paths testable.
- Give operators a stuck-run view.
Watch outs
- In-memory loops break on deploy.
- A generic failed state hides recovery options.
- Human approval delays should not consume worker slots.
Related reading
How this fits the Astra stack
AstraRunner owns the operational path for agent work: orchestration, scheduling, approval gates, traces, costs, and role-agent handoffs. AstraMemory owns durable context and retrieval. AstraGenie turns those capabilities into product workflows for teams that want automation without rebuilding the platform layer.